Skip to main content

Full text of "BART's first five years : transportation and travel impacts : interpretive summary of the final report"

See other formats



DOCUWA l ' • 

Interpretive Summary 
of the Final Report 


This document is disseminated under the 
sponsorship of the U.S. Department of 
Transportation and the U.S. Department 
of Housing and Urban Development in the 
interest of information exchange. The 
United States Government and the Metro- 
politan Transportation Commission as- 
sume no liability for its contents or 
use thereof. 

T*cknico! Report Documenfotion F *5# 

No. 2. Government Accofttien No. 


3. Recipient'* Cotolog No. 

4. T III* end Subtitle 

BART's First Five Years: 

Transportation and Travel Impacts 
Interpretive Summary of the Final Report 

S. Report Do'# 

September 1979 

6 Performing O-gon, tenon Cod* 

6. Performing Orgomtol.on Report No 

D0T-BIP-FR 11-3-78 

7. Author's) 

Alistair Sherret 

9. P •(forming Organ, sot, on Not.* end Address 

Peat, Marwick, Mitchell & Co.,P. 0. Box 8007 
Airport Station, San Francisco, California 94128 
The Metropolitan Transportation Commission 
Hotel Claremont, Berkeley, California 94705 

10. Wort, Unit No. (TRAISj 

Task Order 3 

11. Cent.oct or Gront No. 

D0T-0S-301 76 

13. Type of Report end Period Co»er#d 

Interpretive Summary 
Final Report 

12. Sponsoring Agency Nome and Address 

U.S. Department of Transportation 

and U.S. Department of Housing and Urban Development 

Washington, D.C. 

14. Sponsoring Agency Cod. 


15. Sgpp l#m»ntgry No'* » 

The Metropolitan Transportation Commission is the prime contractor for the 
BART Impact Program. Peat, Marwick, Mitchell & Co. is the subcontractor 
responsible for the Transportation System and Travel Behavior Project. 

16 Abstroet 

BART, the 71-mile San Francisco Bay Area Rapid Transit System, began passenger 
service in 1972. The final section, the transbay link between San Francisco 
and Oakland, opened in 1974. Ridership has grown to about 140,000 passenger 
trips per day, 60,000 of them transbay. This is the final report of a research 
study assessing the impacts of BART on transportation and travel in the Bay Area. 
The BART System, its costs, the service it provides relative to bus and automobile, 
and the nature of its ridership are described. BART's impacts on modal split, 
bus ridership and service, and highway traffic congestion are analyzed; and 
implications for planning rail transit elsewhere are drawn. 

BART's ridership and impacts are less than were widely predicted. This reflects 
both on the optimism of the predictions and the shortcomings of BART's current 
service. As intended, BART's most significant improvements in travel times have 
been for long-distance trips by transit, particularly transbay, to downtown 
San Francisco. Accordingly, the predominant use of BART is for long-distance 
commute trips. BART carries 21% of transbay commute trips. Areawide, BART's 
share of trips for all purposes is between 2% and 3%. Total bus ridership has 
changed little because the loss of riders from service paralleling BART has been 
offset by the use of bus to get to and from BART. Impacts on San Francisco- 
Oakland Bay Bridge traffic have been less than expected because new trips by car 
have appeared to replace those removed by BART. 

17. Key Words 

18. Distribution Stetament 

Bay Area Rapid Transit, BART Impact 
Program, rail rapid transit, urban 
transportation planning, modal split, 
bus ridership, highway traffic. 

Document is available to the public 
through National Technical Information 
Service, Springfield, Virginia 22151 

19. Security Cle**if. {of tKis report) 

20. Security Clost 

si f. (of this peg*) 

21* Ko. of P eg* s 

22. Puce 



Form DOT F 1700.7 («-?2) 

Reproduction of fora; and completed page is authorized 

3 1223 10225 9976 


The BART Impact Program was a comprehensive, policy-oriented study and evaluation 
of tfie impacts of the San Francisco Bay Area's new rapid transit system (BART). 
The program began in 1972, and was completed in 1978. Financing for the Program 
was provided by the U.S. Department of Transportation, the U.S. Department of 
Housing and Urban Development, and the California Department of Transportation. 
Management of the Federally-funded portion of the Program was vested in the U.S. 
Department of Transportation (DOT). The Metropolitan Transportation Commission 
(MTC), a nine-county regional agency established by California law in 1970, 
administered the Program as prime contractor to DOT; the research was performed 
by competitively selected subcontractors to MTC. 

The BART Impact Program studied the broadest feasible range of potential rapid 
transit impacts, including impacts on traffic flow, travel behavior, land use and 
urban development, the environment, the regional economy, social institutions and 
life styles, and public policy. The incidence of these impacts on population 
groups, local areas, and economic sectors was measured and analyzed. 

The results of the BART Impact Program have been synthesized in BART in the Bay 
Area , the BART Impact Program Final Report (PFR). That report was prepared by 
MTC and presents MTC's conclusions from and interpretation of the Program's 
findings. In addition to the PFR, final reports for each of the individual 
projects in the Program were prepared by the consultants who conducted the re- 
search. The reports are listed at the end of this Note. The final reports are 
supported by numerous technical memoranda and working papers. The conclusions 
in those documents reflect the viewpoints of the respective consultants based on 
their research. 

Readers of BART Impact Program reports should be aware of the circumstances and 
the setting in which BART was planned and built and the conditions under which 
the Program was conducted. An understanding of these factors is critical for 
interpreting the Program's findings and attempting to apply them to other areas. 

First, it is important to note that the San Francisco Bay Area has a sound 
economy, a good system of highways and public transportation, and distinctive 
land use and development patterns shaped by the Bay and the hills around it. 

BART was approved and built during a period of vigorous growth in the Bay Area. 
The economy was expanding, suburban development was burgeoning, and major in- 
crements of highway capacity were being added. Also, the Bay Area already had 
extensive public transportation services. There were public carriers operating 
dense networks of local transit services on both sides of the Bay, and there was 
frequent transbay bus service from many parts of the East Bay to San Francisco. 

In 1972 before BART opened, approximately 10% of the total daily trips in the 
three BART counties were made on transit. All of these factors made it difficult 
in the study to isolate BART's effects from other influences that were affecting 
such things as travel behavior and urban development. 

A second important point is that BART was planned and designed primarily to 
facilitate travel from outlying suburbs to downtown areas. Multiple stops are 
provided in the major central business districts, but in other respects BART is 

more like a corrmuter rail system (with long lines and widely-spaced stations) 
than a New York or Chicago-style subway system of interlocking crosstown lines 
and frequent stops. The BART system was intended to rival the automobile in 
comfort, speed, and convenience. Contemporary issues like energy conservation, 
air quality and service for the transportation disadvantaged were not widely 
recognized and publicized concerns during the period of BART's design. 

The institutional setting in the Bay Area was a third important influence on 
BART's development. BART was developed as a separate institution without full 
coordination among existing transportation and regional development planning 
agencies. BART's planners had to make assumptions about policies and develop- 
ment, many of which turned out to be contrary to policies ultimately adopted by 
municipalities in the BART District. 

A critical element in the study design of the BART Impact Program was the defi- 
nition of the No-BART Alternative (NBA), the regional transportation facilities 
and travel patterns judged most likely to have evolved by 1976 if BART had not 
been built. The definition of an NBA was essential since the Program defined 
an impact as the difference between what actually occurred with BART and what 
would have resulted without BART. One cannot be certain about what the region 
would have been like had BART not been built. But based on an analysis of the 
political and economic decision history of the Bay Area and the professional 
judgment of those involved in the Program, it was determined that no significant 
changes to the area's freeway and bridge systems as they actually were in 1976 
would have occurred without BART. It was concluded further that the public transit 
network and services would have been very similar to what they were just before the 
start of BART transbay service. One consequence of this assumption is that the 
NBA provides lower levels of service and less capacity than the with-BART system, 
and attracts fewer riders. The NBA does not extrapolate beyond 1976 and does not 
consider how much additional capacity in the transportation system might eventu- 
ally have been required because of increasing travel demand and congestion. 

An important factor affecting the findings was that BART was not operating at 
its full service level during the period of study by the BART Impact Program. 

The frequency of trains, their operating speeds, the reliability of their oper- 
ations, and the capacities provided in peak periods of travel by BART were 
considerably lower than those originally planned. Trains were running on 
12-minute headways instead of the 4.5 minutes originally planned for each of the 
four lines (90 seconds where three lines converged). BART did not initiate 
service on all lines simultaneously in 1972 but instead phased in service. The 
most critical link, the Transbay Tube, was not opened until late 1974. Night 
service did not start until the end of 1975, and Saturday service started in 
1977. Direct Richmond to Daly City service still is not operating, and it now 
appears that "full service levels," when they are attained, will not achieve the 
headways and average speeds announced in the original plans. 

The final point is that BART had only been operating for a relatively short 
period of time when its impacts were studied. The impact assessment largely 
depends on data collected in the first four years of BART's operations. It is 
likely that some of its impacts, particularly those relating to urban develop- 
ment, will require more time to mature. 

- 2 - 

Final Reports 

These documents are available to the public through the National Technical 
Information Service, Springfield, VA 22151: 

Metropolitan Transportation Commission, "BART in the Bay Area. The Final 
Report of the BART Impact Program," MTC, 1979. 

Gruen Associates, Inc. and DeLeuw, Cather & Company, "Environmental Impacts of 
BART," MTC, 1979. 

Peat, Marwick, Mitchell & Co., "BART's First Five Years: Transportati on and 

Travel Impacts," MTC, 1979. 

Jefferson Associates, Inc., "Impacts of BART on Bay Area Institutions and Life 

Styles," MTC, 1979. 

McDonald & Grefe, Inc., "The Economic and Financial Impacts of BART," 

MTC, 1979. 

John Blayney Associates/David M. Dornbusch & Co., Inc., "Land Use and Urban 
Development Impacts of BART," MTC, 1979. 

Booz, Allen & Hamilton Inc., "The Impact of BART on Public Policy," 

MTC, 1 979. 

Urban Dynamics Associates, "Implications of EART's Impacts for the Transportation 
Di sadvantaged," MTC, 1978. 

Alan M. Voorhees & Associates, Inc., "Federal Policy Implications of BART," 

DOT, 1979. 

- 3 - 

Digitized by the Internet Archive 
in 2015 


BART — the Bay Area Rapid Transit system — is a 
commuter-oriented rail system that serves three counties on 
San Francisco Bay. It began operating in September 1972. 
BART's weekday patronage now is about 146,000 one-way 
trips a day, and the BART staff expects it to reach 180,000 
trips by 1981 . 

In 1972 the Metropolitan Transportation Commission 
began a BART Impact Program - an inquiry into BART's 
effects on people and communities in the region that it 
serves. One element of this program was the Transportation 
System and Travel Behavior (TSTB) Project, whose findings 
are summarized in this report. 

The TSTB Project had three major goals: to describe 
the BART system, its costs, the area that it serves and the 
service that it provides; to analyze BART's patronage and 
the factors that influence it; to assess BART's impact on 
bus service and bus patronage, on highway traffic and 
congestion, and on the division of travel between public 
transit and private automobiles. 

The Project sought an independent assessment of 
BART that would be valuable to transportation planners 

and designers, government officials and the public at large, 
who must make future decisions about transit systems in 
the Bay Area and elsewhere. 

The Project was conducted in two parts. During 
Phase I (finished in October 1975), researchers identified 
BART's initial effects on transportation and travel, then 
collected and analyzed data describing those effects. 
Phase II (finished in December 1977) continued these 
evaluations and analyzed in detail the factors influencing 
commuters' decisions to use or not to use BART. 

Both phases were executed by Peat, Marwick, Mit- 
chell & Co. (San Mateo, Calif.), and were supported by the 
U.S. Department of Transportation and the U.S. Depart- 
ment of Housing and Urban Development. 

The Project focused on transportation and travel 
impacts, and didn't consider BART's effects on urban 
development, land use, the natural environment or the 
social and economic structure of the Bay Area. Those 
effects were studied in other projects of the BART Impact 
Program; they are described in other publications of the 
Program, listed in Section 8 of this report. 


1 The BART Service Area 3 

Definitions of Service Areas 
Other Modes of Transportation 

2 The BART System 4 

BART Fares 

Tickets and Ticket-Processing 
BART Trains 
Train Operations 

3 BART's Costs 7 

Operating Costs and Deficit 

Capital Cost 

4 BART's Impacts on Travel Times and Choices. . . . 8 

Basis for Calculating Travel Times 
BART's Impacts on Travel Times 
Commuters' Comparisons of Modes 

BART Compared with Bus Travel 
BART Compared with Automobile Travel 

5 BART's Patronage and Patrons 10 

Predicted and Actual Patronage 
Sources of Patronage 

Share of Commuting in Primary Service Area 
Potential Share of Commuting Trips 
Share of Travel in the Greater Service Area 
BART's Patrons 

6 BART's Impacts on Other Modes of Travel 13 

Impacts on Transbay Buses 

Impacts on Local Buses 
Impacts on Automobile Traffic 
Impacts on Total Transbay Travel 
Effects on Traffic Peak 

7 Interpretation 15 

Comparing Performance with Expectations 
Future Patronage Growth 

Sources of New Patronage 
Limits on Patronage Growth 
BART's Future Role 

8 Technical Literature 17 


urbanized area 




state highway 

0 1 2 
0 12 3 4 

faderal highway 
interstate highway 

BART station 

1 The BART Service Area 

BART lies within three counties - San Francisco, 
Alameda and Contra Costa - on San Francisco Bay. 
The Bay divides this three-county region into two parts. 
The West Bay comprises the San Francisco Peninsula and 
the City and County of San Francisco. The East Bay in- 
cludes the suburban areas of Alameda and Contra Costa 
Counties, and the older, central cities of Berkeley and 
Oakland (both in Alameda County). 

In both the West Bay and the East Bay, ranges of 
hills have constrained urban development and have defined 
transportation corridors that extend along the shores of 
the Bay. 

BART's four lines radiate from Oakland and are 
named for their termini: 

• The Richmond Line runs north from Oakland, along 
the eastern shore of the Bay, through older, medium- 
density residential and industrial areas. 

• The Fremont Line runs south from Oakland, traver- 
ses medium-density areas, and extends into newer, 
lower-density suburbs. 

• The Concord Line runs through a tunnel in the Berke- 
ley hills and serves newer suburbs in central and 
northern Contra Costa County. 

• The Daly City Line connects the East Bay to San 
Francisco. Its terminus actually is in San Mateo 
County, very close to the southern border of San 

Definitions of Service Areas 

TheTSTB Project defined two BART service areas: 

The primary service area, identified on the map on 
the opposite page, is the area that generates most of 
BART's patronage. About 80% of BART's riders begin 
their trips at points within this area. The population of 
the primary service area in 1975 was about 1.6 million. 

The greater service area comprises San Francisco. 
Alameda and Contra Costa Counties (whose residents pay 
most of the taxes that support BART), plus the northern 
part of San Mateo County (whose residents don't pay 
a direct BART tax). Thus the greater service area embraces 
all of the places that are served - even indirectly - by 
BART. The population of the greater service area in 1975 
was about 2.6 million. 

Other Modes of Transportation 

The major transportation resources in BART's 
greater service area are BART itself, highways in all of the 
important transportation corridors, and local transit 
systems that operate buses, streetcars or both. 

All of BART's lines are in established transportation 
corridors and are roughly parallel to major highways. For 
assessing BART's impact on automobile traffic, the most 
important highway section is the part of Interstate Highway 
80 that traverses the Bay on the San Francisco-Oakland Bay 
Bridge. Other major highways that are parallel to ele- 
ments of BART include: Interstate Highway 280 (connect- 
ing San Francisco to San Mateo County); California High- 
way 17 (along BART's Richmond and Fremont Lines): 
and California Highway 24 (near BART's Concord Line), 
which carries commuters between the Bay Bridge and 
some distant suburbs. 

The most important public transit agencies in 
BART's greater service area are the San Francisco Municipal 
Railway (MUNI) and the Alameda-Contra Costa Transit 
District (AC Transit). MUNI operates buses, streetcars, 
trolley cars and cable cars within San Francisco. AC Transit 
runs local and express buses in the East Bay, and express 
buses to San Francisco from Oakland, Berkeley and several 
other communities. 

Many of MUNI's and AC Transit's local buses also 
serve BART stations, and they can be regarded as ex- 
tensions of the BART system. This is especially true of 
buses that AC Transit operates, under a contract with 
BART, in outlying parts of BART's greater service area. 

Map on the opposite page: All four of BART's lines are 
in established transportation corridors and are roughly 
parallel to major Bay Area highways. Approximately 80% 
of BART's riders begin their journeys at points within the 
primary service area, which is shown on this map by light blue. 


2 The BART System 

The basic concepts and standards for BART appeared 
in the Composite Report* that the BART District received 
in 1962 from its engineering and financial consultants. The 
report described BART as a regional rapid-transit system 
whose speed, comfort and convenient service would make 
it competitive with the automobile - especally for peak- 
period travel from the more remote suburbs to the central 
business districts of San Francisco and Oakland. By attract- 
ing automobile commuters from their cars, the report said, 
BART would relieve "the continuing increase of traffic 
congestion which threatens the future growth and well- 
being of the San Francisco Bay Area." The system wouldn't 
be notably useful for off-peak travel or for short trips 
within the central cities; about 75% of BART's patrons 
would be commuting between their homes and their jobs. 

The desire to attract long-distance commuters who 
otherwise would use private automobiles was translated 
into major design objectives for the system: 

• BART trains would be able to accelerate quickly and 
to maintain a cruising speed of 70 miles an hour, at 

• Suburban stations would be rather far apart, so that 
the trains could exploit their high speed. The average 
speed, including station stops, would be 45 miles 
an hour. 

• BART cars would be light, to facilitate rapid ac- 
celeration and decelertion, and to minimize their 
consumption of electric power. 

• BART's tracks would be broad-gaged (the rails would 
be 5’/2 feet apart), so that the light cars would be 
stable at high speeds. 

• Trains and stations would be comfortable and at- 

• During peak-travel periods, the time ("headway") 
between successive trains on the busiest parts of the 
system would be as little as 1 'A minutes. 

• Trains would be dispatched and routed by automatic 
equipment, so that they could run safely with such 
short headways. 

These objectives eventually were met, with two ex- 
ceptions: The average speed of BART trains is about 38 
miles an hour, not 45; and minimum headways are about 
6 minutes, not 1%. BART now is trying to diminish its 
minimum headways to about 3 minutes. 

•Parsons Brinckerhoff-Tudor-Bechtel, Smith Barney & Co., Stone & 
Youngberg, and Van Beuren Stanbery. 1962. The Composite 
Report, Bay Area Rapid Transit. 


BART's tracks are about 71 miles long. About 20 
miles of tracks are underground: 13 miles in subway tun- 
nels built by boring or by cut-and-cover work; 4 miles in 
the Transbay Tube beneath San Francisco Bay; and 3 miles 
in the bored tunnel that carries the Concord Line through 
the Berkeley hills (map, page 4). Of the remaining 51 miles 
of BART tracks, 24 miles are on eleveted structures while 
27 miles are at ground level or on earth embankments. 

About 85% of BART's trackways lie within, beside 
or below the rights-of-way of arterial streets, highways or 
other railroads. 


BART has 34 stations, of which 14 are below ground. 
Stations in downtown San Francisco and Oakland are 0.3 
to 0.5 miles apart. Suburban stations are 2 to 4 miles apart. 

There are parking lots at 23 of the stations, and their 
capacities range from 240 to 1,600 cars. The combined 
capacity of the 23 lots is about 20,000 cars. 

BART's stations display high standards of design and 
construction. They reflect the assumption by BART's 
planners that a pleasant environment for traveling would be 
important for attracting motorists out of their automobiles 
and into rapid-transit trains. 

Each station has two or three levels — a concourse 
and one or two platform levels. Patrons enter and leave 
through the concourse, which houses a station agent, 
ticket-selling machines, faregates, restrooms, telephones 
and candy-vending machines.* Patrons board their trains 
on the platforms, which are connected to the concourse 
by stairs, by escalators, and by elevators that can be used 
by handicapped persons. 

The stations were designed by 15 different architec- 
tural firms. BART established engineering standards for the 
stations, but it imposed few restrictions on style, appoint- 
ments, finish materials or other architectural features. As 
a result, station concourses differ in size and layout, and 
their individuality is amplified by distinctive materials and 
works of art. Advertising displays generally are restrained. 
BART stations are pleasingly free of graffiti and other 
marks of vandalism. This can be ascribed to several factors: 

A few stations also have shops on their concourses, and some 
downtown stations have direct connections to adjacent stores 
and office buildings. 


The concourse of BART's Berkeley station. The high design 
and construction standards seen in BART stations reflect the 
assumption that a pleasant environment would attract mo- 
torists out of their automobiles and into rapid-transit trains. 

the extensive use of ceramic tile (and other unreceptive 
materials) in many stations; diligent maintenance; and the 
spacious, open design of station interiors, which facilitates 
continual surveillance by BART personnel and by tele- 
vision cameras. 

BART Fares 

BART's fares range from $0.25 to $1.45. The chief 
determinant of a fare is the distance travelled. Fares from 
downtown San Francisco to some of the destinations 
shown on the map on page 2 are: 

$0.55 to Daly City $1 .20 to Lafayette 

0.75 to Oakland 1 .35 to Concord 

1 .00 to Richmond 1 .40 to Fremont 

Patrons aged 65 or older pay only 10% of the normal 
fare for any trip. Handicapped persons and children aged 
5 to 12 pay only 25%. Children under 5 pay nothing. 

Tickets and Ticket-Processing 

The full-fare tickets used by most passengers are 
sold from machines on station concourses, in denomina- 
tions from $0.25 to $20.00. The value left in a ticket after 
a trip can be used for later trips. 

The discounted tickets used by the old, the young 
and the handicapped aren't sold at stations. They can be 
bought at local banks only. 

Tickets are processed by automatic faregates at both 
the origin and the destination of each trip. At the origin 
station, the passenger inserts the ticket into a faregate 
that marks it magnetically with an origin code and returns 
it to the passenger. At the destination station, another 
faregate reads the origin code, subtracts the applicable 
fare from the value of the ticket, marks the ticket to show 
the value remaining in it, and returns the ticket to the 


BART Trains 

The passenger section of a BART car is 70 feet long. 
Each car can carry 72 seated passengers. BART's designers 
expected its trains to run so frequently that all of its pa- 
trons would ride seated, so they provided seat handles but 
no grab rails for standing passengers. In practice, the peak- 
period trains on some lines carry as many passengers stand- 
ing as seated, and BART has installed overhead grab rails 
in all of its cars. 

The cars are unusually attractive and comfortable. 
Their interior features include air-conditioning, carpeting, 
tinted window glass and upholstered seats. These appoint- 
ments reflect again BART's hope that pleasing surroundings 
would attract commuters out of their automobiles. 

An operator in the first car of each train monitors 
the performance of the train's equipment and its respon- 
ses to commands from the automatic train-control (ATC) 
system that dispatches trains and controls their speeds, 
spacings and stops. In certain situations, the operator can 
override some of the train's automatic functions and 
assume partial manual control. Each car has a commun- 
ication system that the operator uses to announce stations 
and transfer points. Passengers can speak to the operator 
through an intercom system. 

Train length varies between 3 and 10 cars. During 
peak-travel periods, most of the trains on all four lines 
have 10 cars. The trains can reach 80 miles an hour, but 
their maximum speed in ordinary service is 70. 

The comfort of a BART ride is outstanding. BART 
cars run with little vibration or sway, for four important 
reasons: BART tracks are broad-gaged; they are continuous, 
welded rails; they are well maintained; and each car is sup- 
ported by pneumatic cushions, which are more effective 
than springs in absorbing vibration. 

Sound levels in BART cars are low. Patrons hear 
less noise than they would hear in the cars of most other 
systems, and they can converse with companions in a 
normal voice. 

Train Operations 

BART began service between Oakland and Fremont 
in September 1972, between Richmond and Fremont in 
January 1973; between Concord and Oakland in May 1973; 
and between Daly City and downtown San Francisco in 
November 1973. Trains began running through the Trans- 
bay Tube in September 1974 - between Concord and Daly 
City, and between Fremont and Daly City. 

Initial service over all of these routes was confined to 
weekdays, between 6:00 AM and 8:00 PM. BART began 

nighttime service in November 1975, with trains running 
until midnight. Saturday service (6:00 AM to midnight) 
began in November 1977, and Sunday service (9:00 AM to 
midnight) in July 1978. 

BART's average scheduled headways today are: 12 

minutes on weekdays between 6:00 AM and 6:00 PM; 
20 minutes on weekdays between 6:00 PM and midnight; 
and 20 minutes on Saturdays and Sundays, throughout 
the day. 

BART plans to expand its service significantly during 
the next two years by increasing the frequency of trains 
on all lines during peak travel periods, and by starting 
direct service from Richmond to San Francisco and Daly 
City. (Travelers from Richmond to San Francisco or 
Daly City now must transfer between trains at either the 
MacArthur or the 12th Street station in Oakland). 


From the outset, BART has had serious problems with 
its equipment. During the 12-month period ended in June 
1978, about 10 trains a day suffered failures that forced 
BART to remove them from service. In about half of 
these cases the train was able to finish its run before it 
was withdrawn. In the rest, passengers had to leave the 
affected train and finish their journeys on another. 

As a result of such incidents, BART carries a repu- 
tation for unreliability — a reputation that doubtless has 
deterred some people from riding its trains. 

BART's most important operating problems arise 
when it must deal with a malfunctioning train or a train 
that has a malfunctioning car. Such a train can't be oper- 
ated at normal speeds; and it usually can't be diverted 
promptly, because BART has relatively few sidings, yards 
or crossovers. Hence it must travel on a main track for a 
considerable distance, at a reduced speed, delaying all 
the trains that are behind it. 

These difficulties are worsened by restrictions on 
BART's use of the ATC system. BART won't be able to 
exploit this system fully until it is approved by Califor- 
nia's Public Utilities Commission, which must certify the 
safety of BART's operations. 

BART already has taken some important steps 
toward improving the handling of disabled trains. It has 
built new sidings, it has planned an additional track 
through its busiest segment (the subway tunnel in Oak- 
land), and it has developed new procedures that will enable 
trains with certain kinds of malfunctions to run at normal 
speeds until they can be diverted to a yard. BART will 
begin using these new procedures when they are approved 
by the Public Utilities Commission. 


3 BART's Costs 

Operating Cost and Deficit 

During the year ended on June 30, 1978, BART's 
revenue from fares was about $28 million and its total 
operating cost was some $78 million. About 80% of the 
resulting deficit was paid from a sales tax (V&cf per dollar) 
and about 10% from a property tax (5tf per hundred 
dollars of assessed value) collected in the three BART 
counties only. 

BART's farebox ratio - the percentage of operating 
cost that was covered by fares - was about 35%. This was 
slightly higher than the ratios reported by MUNI and AC 

BART collected $0.73 and spent $2.02 for each 
passenger carried. Its expense for each passenger seems 
higher than corresponding outlays by other rail systems 
in the United States. But such comparisons can be mis- 
leading because the average trip on BART is almost 13 
miles long - much longer than the average trip on the rail 
rapid-transit systems in New York, Chicago and Phila- 

BART's operating expense for each passenger-mile - 
a quantity that compensates for differences in average trip 
length - was about 15tf. This falls within the range of ex- 
pense ( 1 0<^ to 15d) reported by other systems. 

BART's expenses during the year ended on June 30, 1978. 

Total spending was $90.4 million. Operating expenses ac- 
counted for $78.2 million (87%) of this. The BART Dis- 
trict's operating deficit for the year was $50.0 million. 

Capital Cost 

The Composite Report of 1962 said that BART 
would cost about $994 million, and that the major ele- 
ments of the system could be built in about six years 
(between mid-1963 and the end of 1968). 

Major construction didn't begin until mid-1964 and 
wasn't completed until September 1974, when the Trans- 
bay Tube opened for service. Minor construction work is 
continuing still. 

In March 1977, BART's actual capital cost had reached 
$1,554 million. According to a BART report issued in 
1975, the cost will exceed $1,636 million by the end of 
1980. (About 10% of this will be the cost of modern- 
izing San Francisco streetcar lines that will share certain 
underground routes and stations with BART). The dif- 
ference between the cost predicted and the cost now 
foreseen is $642 million. This excess is due to three major 

• Changes in BART's scope and design Between 1962 
and 1975, the plans for BART underwent substantial 
changes. Some of these changes reduced BART's cap- 
ital cost, others increased it. Their net effect was to 
increase it by $159 million (25% of the excess cited 

Revenue sources during the year ended on June 30, 1978. 
BART's total income was $90.4 million. Local property 
taxes and a local sales tax provided 60.8% of this, and 
paid for about 90% of the District's operating deficit. 


Changes that reduced capital expenses included the 
elimination of one track from the BART tunnel be- 
neath Oakland, and a reduction in the number of park- 
ing spaces at BART-station parking lots. Plans pro- 
posed^ in 1962 contemplated a total of 36,000 spaces in 
the lots. But that number was reduced to 24,000 in 
1963, after a new estimate of parking demand had 
been made. And it later was reduced to 18,000 to cut 
constructions costs. 

Changes that increased capital cost included the re- 
designing of some stations; the addition of a station 
(Embarcadero, in San Francisco) that wasn't included 
in the 1962 plan; the installation of facilities for hand- 
icapped patrons; and the placing of all Berkeley tracks 
in subway tunnels. (The 1962 plan had assumed that 
only the tracks in central Berkeley would be under- 
ground, while the tracks in outer districts would be 
at ground level). 

Unexpectedly high inflation rates The Composite 
Report assumed that the average annual inflation rate 
during the six-year construction period (1963 through 
1968) would be 3%. The actual rate during those years 
was 5.7%. Even if construction had been finished by 
the end of 1968, this unforeseen inflation would have 
expanded BART's capital cost by $94 million (14% of 
the excess). 

• Construction delays The major sources of delays were 
design changes, problems with financing, and a tax- 
payer's lawsuit challenging the 1962 election in which 
the BART bond issue had been approved. These delays 
added $321 million (50% of the excess) to BART's 
cost, by extending construction into years of fur- 
ther and faster inflation. Between December 1968 
(when the Transbay Tube originally was scheduled for 
completion) and September 1974 (when the Tube 
actually opened for service), the average annual in- 
flation rate was 1 1 .3%. The average rate during BART's 
entire 1 1-year construction period was 8.3%. 

Together, these three factors accounted for $574 
million (89%) of the $642 million excess. The remaining 
$68 million is due to factors that can't be assigned to any 
of the three foregoing categories. 

BART's capital expenditures (unlike those of most 
other rail rapid-transit systems now planned or under 
construction in the United States) are being paid chiefly 
from local funds. If BART's capital cost does indeed 
reach $1,636 million in 1980, this amount will have been 
raised in the following ways: $942 million from bonds 
serviced through a local sales tax and a property tax; 
$325 million from federal grants; $176 million from tolls 
collected on transbay bridges; and $181 million from 
interest and other local sources. 

4 BART's Impacts on Travel Times and Choices 

The TSTB Project used two tactics for assessing 
BART's impacts on transportation service and travel times: 
calculating the times that would be required for hypotheti- 
cal trips in the BART greater service area; and inter- 
viewing commuters who could make their trips by BART 
or by another mode of transportation. 

Basis for Calculating Travel Times 

BART's aggregate effects on travel times were ana- 
lyzed by calculating the time required to make hypothetical 
trips in three different ways: 

• using the with-BART public-transportation system - 
an approximation of the transit network, including 
BART, that actually existed in 1976; 

• using a hypothetical no-BART public-transportation 
system — an approximation of the transit network 
that might have existed in 1976 if BART hadn't 
been built; 

• using hypothetical travel by automobile, over a net- 
work of highways and roads that approximated those 
that actually existed in 1976. 

The hypothetical no-BART system was developed 
because a simple comparison of conditions in 1962 (before 
BART was built) and 1976 (after BART was built) would 
have been misleading. The Bay Area's public-transit net- 
work wouldn't have remained unchanged for 14 years if 
the voters hadn't approved BART. 

A careful historical study of economic and politi- 
cal decisions suggested that, if BART hadn't been built, 
the Bay Area would have had in 1976 a public-transit 
system much like the one that actually was operating there 
in 1973. Hence the 1973 system, with some minor changes, 
was used as the no-BART system. This system, which con- 
sisted chiefly of local and express buses, would have pro- 
vided less service and less capacity than the with-BART 
system provided in 1976; and it would have attracted 
fewer patrons. 

BART's Impacts on Travel Times 

During peak-travel periods, the with-BART system 
was faster than the no-BART system for most of the trips 
that were studied. For .travel to 50 major centers of em- 


ployment in the greater service area, the average with- 
BART trip was 5 minutes (12%) shorter than the average 
no-BART trip. The with-BART advantage was greatest for 
trips from outlying suburbs to the destinations in central 
Oakland, Berkeley and San Francisco. 

But with-BART travel was generally and markedly 
slower than travel by automobile. For journeys to the same 
50 centers of employment, the average automobile trip was 
14 minutes (34%) shorter than the average with-BART trip. 
The times required for with-BART travel and for automobile 
travel were comparable only for long trips between distant 
suburbs and central Oakland, Berkeley and San Francisco. 

These findings demonstrate that BART is most ef- 
fective when it is serving its primary purpose, stated in 
1962: carrying commuters on long trips between their 
homes in the suburbs and their jobs in urban centers. The 
findings also suggest how commuters might respond to 
BART if they relied entirely on true travel time to dictate 
a mode of transportation. 

But commuters actually consider several different 
factors when they decide how to travel — factors such 
as travel time, comfort, cost and convenience. And their 
choices are influenced by subjective evaluations of these 
variables, whether accurate or not. To study commuters' 
evaluations of BART, and to learn how these evaluations 
influence patronage, the TSTB Project conducted the 
survey described next. 

Commuters' Comparisons of Modes 

In 1977, the Project asked 2,257 peak-period com- 
muters to compare BART with an alternative mode of 
transportation. Each of these respondents both lived and 
worked within BART's primary service area, and each 
could commute by BART or by another mode: 625 said 
that they could use BART or bus, and 1,632 said they could 
travel by BART or auto. The survey questionnaire asked 
respondents to assign importance to factors such as travel 
time, cost, comfort, speed and safety, and to evaluate 
BART against its alternative with respect to each factor. 

BART Compared with Bus Travel Among commuters 
who could travel by BART or by bus, 54% actually used 
BART and 46% used buses. 

Those who actually used BART said that the most 
important aspects of a commuting trip were total travel 
time, convenience, and the amount of time spent waiting 
for a vehicle or transferring between vehicles. The average 
trip made by these commuters took 46 minutes by BART 
and cost $1.34. If the same trips had been made by bus, 
the BART-riders thought, the average trip would have been 
11 minutes (24%) longer and would have cost $0.93 
for fares. 

The importance of travel time in BART-or-bus choices. 
For commuters who rode BART (upper two bars), the average 
trip took 46 minutes; their average bus trip would have 
taken 57 minutes. The lower two bars show the correspond- 
ing averages for commuters who rode buses instead of BART. 

Those who actually rode buses said that they were 
interested chiefly in total travel cost, convenience and re- 
liability. The average trip made by these respondents took 
46 minutes by bus and cost $0.97. If the same trips had 
been made by BART, they thought, the average trip 
would have been 5 minutes (11%) longer and would have 
cost $1 .44. 

BART Compared with Automobile Travel Among commu- 
ters who could choose BART or automobile, 27% actually 
used BART and 73% drove alone in automobiles. (Com- 
muters who rode in car pools weren't considered in this 

Those who actually used BART said that the most 
important aspects of a trip were total cost and the stresses 
of driving an automobile in peak-period traffic. The average 
trip by these respondents took 52 minutes by BART and 
cost $1.62 in fares. For corresponding trips by automobile, 
the average trip would have been 18 minutes (35%) shorter, 
and would have cost $2.55 for tolls and parking. 

Those who usually drove alone in automobiles 
thought that the most important aspects of a trip were 
convenience, total travel time and reliability. The average 
trip made by these commuters took 27 minutes by auto- 
mobile and cost $1.83 for tolls and parking fees. If the 
same trips had been made by BART, they believed, the 
average trip would have been 28 minutes (104%) longer, 
and would cost $1 .58 in fares. 

Analysis When BART is faster than a bus, BART draws 
riders despite its higher cost. But BART rarely is faster than 
an automobile; and despite the higher cost of using an auto, 
BART attracts far fewer riders from autos than from buses. 
Those commuters who choose BART over an automobile 
accept BART's longer travel times and unreliability because 
they don't like the high cost and the stress of driving. 
This suggests that BART might attract more auto-users 
in the future if BART's reliability and headways improve, 
and if the costs and stresses of auto travel increase. 


5 BART's Patronage and Patrons 

Predicted and Actual Patronage 

BART's patronage* hasn't reached predicted values, 
chiefly because BART isn't providing all of the service that 
was predicted. 

The earliest forecast of patronage appeared in the 
Composite Report of 1962: BART's weekday patronage 
would reach 259,000 by 1975 - four years after all parts 
of the system were to be opened for service. Lower figures 
appeared in later documents to compensate for delays in 
construction, in the opening of new segments of the 
system, and in the introduction of nighttime service. 
BART's most recent estimate of weekday patronage is 

180,000 in 1981, after all of the service expansions now 
planned have been made. 

In March 1975, six months after BART began service 
through the Transbay Tube, average weekday patronage 
was 116,000. It has grown modestly since then. In March 
1978 it was 144,000. In September 1978, four years after 
all parts of the system had been opened for service, it was 
146,000. The growth in patronage hasn't been evenly 
distributed throughout the system. For example: In March 

•Throughout this report, patronage means one-way trips a day. 

1975, the average combined weekday patronage on the 
three East Bay lines was 38,000, by March 1978, it had 
increased by only 2,000. During the same three years, 
however, transbay patronage on weekdays grew from 

51.000 to 63,000. 

BART's average Saturday patronage during the 12 
months ended on December 31, 1978, was about 60,000. 
Average Sunday patronage during the six months ended 
on the same date was about 35,000. 

Sources of Patronage 

About 70% of the BART trips made on weekdays 
are commuting trips to work, and about 10% are commut- 
ing trips to colleges and other schools. Commuters to work 
travel chiefly to downtown Oakland or San Francisco, 
from outlying stations on the Daly City and Concord 
Lines. Commuters to schools travel chiefly to colleges and 
universities in Berkeley, Oakland and Hayward. The busiest 
segment of the system is the Transbay Tube, which handles 

30.000 trips a day in each direction. Travel through the 
Tube (and through most other parts of the system) is 
highly directional: During the morning peak-travel period, 
about 90% of the people who ride through the Tube 
are moving toward San Francisco. During the evening 
peak-travel period, the opposite is true. 


The average trip on BART is 13 miles long and costs 
$0.75. The average travel time is 46 minutes; this includes 
20 minutes spent in getting to and from BART stations 
and waiting for a train. About 60% of BART's patrons 
ride to their stations in automobiles — 40% as drivers, 20% 
as passengers. Of the rest, about half walk to their stations 
and about half use buses. 

Among patrons who use BART for trips that they 
formerly made by some other mode, about 45% formerly 
rode buses and 55% formerly traveled in automobiles. 
The percentage for former auto-users is higher than that 
reported for new rapid-transit systems elsewhere. For 
example: On the Lindenwold line near Philadelphia, which 
serves a suburban area similar to some of BART's corridors, 
the corresponding figure is 39%*. 

Share of Commuting in the Primary Service Area 

BART's share of commuting trips to workplaces in 
its primary service area was estimated from a 1977 survey 
of 8,400 commuters. 

About 13% of these commuters used BART, 17% 
used buses, 61% used automobiles and 9% used other 
modes. BART handled about 17% of the trips to work- 
places in downtown San Francisco and 25% of the trips 
to workplaces in downtown Oakland. These are percent- 

‘American Automobile Association. 1971. The Lindenwold Line: 
A Case Study of the Newest Rail Rapid Transit. 

BART's importance in the transbay corridor. The chart at 
the left shows the modes of transportation that people use 
for traveling on weekdays in BART's greater service area. 
BART's share of this travel is only 2%. But its share of 

ages for the entire group — those whose homes lay in- 
side the primary service area, and those who commuted 
into the primary service area from homes outside it. 

Among the 5,000 respondents whose workplaces 
and homes were inside the primary service area, about 
16% commuted by BART while 14% used buses, 59% 
used autos and 11% used other modes. BART handled 
about 23% of the trips to downtown San Francisco and 
about 27% of the trips to downtown Oakland from homes 
in the primary service area. 

Potential Share of Commuting Trips 

Respondents who both lived and worked in BART's 
primary service area said that BART was a practical mode 
of travel for 40% of their trips between their homes and 
their jobs. But they actually were using BART for only 16% 
of their trips. This implies that BART can increase its 
share of commuting trips significantly, if it can improve its 
service, speed and reliability. 

Share of Travel in the Greater Service Area 

A survey was made in 1975 to analyze weekday 
travel in BART's greater service area, by purpose and by 
mode of transportation. It showed that BART is an im- 
portant resource for commuting, but not for most other 
kinds of travel. This demonstrates again that BART is 
serving the one important purpose that was proposed for it 
in 1962: carrying long-distance commuters. 

commuting trips in the transbay corridor is 32%, as shown 
by the chart at the right. These charts are based on two 
studies that were conducted by the TSTB Project: the 1975 
Areawide Travel Survey and the 1977 Work Travel Survey. 


Surveys made in 1975 and 1977 found that BART 
handled only 2.4% of all weekday trips in the greater ser- 
vice area, but it carried 3.8% of the trips made during 
peak-travel periods, 5.2% of all commuting trips and 32% 
of commuting trips in the transbay corridor (as shown in 
the right-hand graph on page 11). These percentages and 
the ones given earlier for BART's share of commuting in 
the primary service area demonstrate that BART's impor- 
tance in a travel market depends strongly on the definition 
of that market. 

BART is more attractive for long trips than for short 
ones. The time spent in traveling to a BART station is onfy 
a small fraction of the total time required for a long trip, 
but it is a substantial part of the time required for a short 
trip. BART handles less than 1% of the trips that take less 
than 16 minutes; but it carries 13% of the trips that take 
56 to 75 minutes, and 20% of the rare trips that take more 
than 75 minutes. This attractiveness for relatively long com- 
muting trips is a feature of buses, too, for travel in the 
greater service area. For short trips, public-transit vehicles 
rarely can compete with the speed and convenience of the 
automobile, except in the dense, congested areas of San 
Francisco. There, parking and traffic conditions can make 
public transit a more attractive choice for many short 

BART's Patrons 

The 1976 Passenger Profile Survey — a study of about 
8,000 BART-riders, conducted jointly by MTC and the 
BART District - showed that BART's adult patrons are 

than to to to or 

10,000 14,999 19,999 24,999 more 

annual household income (dollars) 

Comparing BART patrons with the population at large. 
The chart at the left shows that BART's adult patrons (white 
bars) come from households whose annual incomes are some- 
what higher than household incomes in the genera / popu- 

much better educated and somewhat younger than the 
general population of the greater service area. They include 
a slightly higher percentage of males, and they represent 
households whose annual incomes are somewhat higher 
than household incomes among the general population. 

The percentage of black persons (11.4%) among 
BART's patrons is comparable to their representation in 
the general population (11.2%). The same is true of Asians 
and other minority groups, excepting Spanish-Americans. 
Spanish-Americans constitute about 11% of the people in 
the greater service area, but only 6% of BART's riders. This 
might be true because the largest concentration of Spanish- 
Americans lies in San Francisco's Mission District, which 
enjoys excellent bus and trolley-car service. 

BART was the first public-transit system in the 
United States that was designed to accommodate handi- 
capped travelers, including travelers in wheelchairs. None- 
theless, the percentage of handicapped persons among 
BART's patrons is much smaller than their percentage in 
the general population. 

Handicapped persons interviewed during the TSTB 
Project said that BART's trains and stations still don't pro- 
vide some of the special features that th^ handicapped re- 
quire. But two other facts seem more important in ex- 
plaining why relatively few handicapped persons use 
BART: They have difficulty in reaching BART stations by 
bus or by automobile. And they generally don't perceive 
themselves as being mobile. As a result, they do less travel- 
ing, by any mode, than other people do. 

lation (black bars) of BART's greater service area. The chart 
at the right shows that BART patrons have had considerably 
more education. Both of these illustrations reflect data gath- 
ered by BART during its 1976 Passenger Profile Survey. 


6 BART's Impacts on Other Modes of T ravel 

About 27% of the BART-riders interviewed during 
the 1976 Passenger Profile Survey said that they were 
using BART for trips that they formerly hadn't made at 
all. They had begun riding BART after they had acquired 
new jobs or new homes. The remaining 73% were (in about 
equal numbers) former bus-riders or former automobile- 
users. The diversion of these travelers to BART has had 
some impacts on transbay and local buses, but no lasting 
effect on automobile traffic. 

Impacts on Transbay Buses 

BART's greatest effect on bus patronage has occurred 
in the transbay travel corridor. In February 1978, BART 
was handling about 60,000 transbay trips a day on week- 
days. More than half of these trips had been diverted from 
AC Transit and Greyhound buses. 

During the first three months of 1974, the average 
weekday patronage on AC Transit's transbay buses was 
63,000 trips a day. During the first three months of 1975, 
after BART had begun service through the Transbay Tube, 
AC Transit's average transbay patronage was 44,200 - 
about 30% less than it had been in 1974. During the first 
three months of 1976, patronage rose to 45,200. As BART 
diverted travelers from transbay buses, AC Transit gradually 
reduced its transbay service. In December 1973, AC Trans- 
it’s transbay lines accounted for 35,800 bus-miles a day 
on weekdays. By December 1976, service had been cut 
to 30,400 bus-miles. 

BART's effects on Greyhound were even greater. 
Under a plan approved by the Public Utilities Commission, 
Greyhound curtailed its service sharply on lines that link 
San Francisco to Oakland and to Contra Costa County, 
after BART began to draw riders from these lines. Grey- 
hound had wanted for some years to reduce its commuter- 
bus service on these routes, but the PUC had required the 
company to operate a substantial number of such buses 
until BART could provide commuters with an alternative 
mode of travel. 

In March 1974, Greyhound buses were making 242 
trips a day on weekdays and carrying 12,200 passengers. 
By the end of 1976, they were making only 42 trips a day 
and carrying about 1,700 passengers. The reduction in bus 
trips was about 80% — approximately equal to the re- 
duction in patronage. 

Greyhound also has discontinued all of its weekday 
off-peak service on routes that compete with BART's 
Concord Line, although the total number of off-peak trave- 
lers in the Concord travel corridor has increased substan- 
tially. Most of the increase has been absorbed by BART. 

Impacts on Local Buses 

Local bus lines interact with BART in two ways. 
They compete with BART (thus losing some patronage to 
BART) for long commuting trips. And they complement. 
BART (thus gaining some patronage) by moving travelers 
to and from BART stations. 

On both sides of the Bay, these two kinds of inter- 
action have nearly offset each other. BART has had no 
significant effect on the total patronage of local buses 
operated by MUNI in San Francisco or by AC Transit 
in the East Bay. But BART obviously has reduced or 
increased patronage on certain of the local bus lines opera- 
ted by each of these agencies. 

AC Transit, MUNI and some smaller bus operators 
have reduced service on some local lines that compete with 
BART, and have expanded service on some lines that carry 
patrons to BART stations. And in outlying parts of the 
greater service area, AC Transit is operating BART express 
buses that not only serve BART stations but also supple- 
ment local public-transit service. 

In general, however, BART has produced only small 
effects on the nature and frequency of local bus service. 
There are two reasons for this. First: The local buses are 
operated by several independent agencies — not by a 
single authority that could have optimized service changes 
in response to BART. Second: Some of the local service 
that these agencies wanted to eliminate was retained 
because patrons protested the planned curtailments. 

Impacts on Automobile Traffic 

BART has had little effect on the volume or speed of 
automobile traffic, or on its distribution through the day. 
This finding is based on analyses of actual and predicted 
traffic volumes, between January 1972 and October 1976, 
on the four major bridges that cross the Bay (map, page 4). 
Of these, the Bay Bridge alone was affected by BART. 
The three other bridges provided baselines for the analysis 
of BART's impacts. 

Impacts on Total Transbay Travel 

Since BART began transbay service, the total capacity 
of the transbay travel corridor has increased substantially. 
So have the number of trips actually made in the corridor 
and the percentage of trips made in transit vehicles (BART 
and buses). 

The five lanes of the Bay Bridge can handle about 
8,700 automobiles an hour. During peak periods, the 


average occupancy rate for autos crossing the bridge is 1.6, 
so the bridge can carry about 14,000 passengers an hour. 

With 10-car trains running through the Transbay Tube 
at 6-minute intervals, BART can carry about 8,600 passen- 
gers an hour: 7,200 passengers seated and 1,400 stand- 
ing - a typical ratio during peak periods. Hence BART's 
transbay capacity is about 61% of the bridge's capacity. 

The total number of transbay trips actually made has 
increased by about 27% during the past five years - from 
112,400 trips a day in 1973, to 142,500 trips a day in 
1977-1978*. Most of this increase has been absorbed by 
BART, as the graph on this page shows. 

During the same period, the percentage of transbay 
trips made in public-transit vehicles grew from 25% to 33%, 
while the percentage of trips made in private automobiles 
decreased accordingly. 

Thus BART has had two important impacts in the 
transbay corridor: It has increased the corridor's capacity, 
and it has absorbed most of the increase in commuting 
trips through the corridor. It hasn't, however, reduced the 
number of vehicles that use the Bay Bridge. 

During June 1974, before BART began service through 
the Transbay Tube, the bridge handled about 93,000 
vehicles a day, in each direction, on weekdays. 

Immediately after the Tube was opened, traffic on 
the bridge declined noticeably. By December 1974 it had 
fallen to 85,000 vehicles. 

But then it began to recover. In June 1975 it was 
94,000; and in October 1976 it reached 96,000 — almost 
exactly the volume that would have been expected if there 
had been no transbay BART service. 

The source of this resurgence of traffic can't be identi- 
fied with certainty. It might be attributable to local popu- 
lation growth or economic changes during the period under 
study. It might represent trips that were captured from 
other routes when BART temporarily relieved congestion 
on the bridge. Or it might represent induced travel - trips 
that previously had been discouraged by congestion on the 
bridge, but that became attractive as soon as BART began 
to relieve that congestion. 

Induced travel is a common phenomenon. Wherever an 
automobile route is used heavily, there exists a reservoir 
of trips that people don't make because the route can't 

The figure given for 1973 is the average number of transbay trips 
made on weekdays in April and October 1973. The figure for 
1977-1978 is the average number of weekday trips in October 
1977 and April 1978. 

BART's importance in absorbing new transbay travel. Average 
weekday travel in the transbay corridor has grown from 
1 12,400 trips in 1973 t the upper bar) to 142,500 trips in 
1977-1978 (the lower bar). The difference is 30,100 trips. 
Nearly all of this new travel has been absorbed by BART. 

accommodate them. If a second route is provided, both 
routes will draw traffic from this reservoir; and the net 
loss in traffic by the old route will be considerably less 
than the gain in traffic by the new route. This result some- 
times is surprising to officials and to the public, who 
reason that the construction of new transportation facil- 
ities (such as BART) must substantially reduce the load 
on others. 

BART's impact on Bay Bridge traffic might have been 
diluted by some changes in the operation of the bridge 
itself. These changes, introduced between December 1971 
and May 1975, included the reservation of certain toll 
lanes for buses and car-pools, the elimination of peak- 
period tolls for automobiles carrying three persons or more, 
and the installation of a signal system for optimizing the 
flow of traffic. 

Effects on Traffic Peak 

After BART began transbay operations, the morning 
peak-traffic period on the bridge became slightly shorter, 
and its center shifted from about 7:30 AM to 7:45 AM. 

A similar shift was observed on California Highway 24, 
an important artery for vehicles moving toward the bridge. 
Again there was no significant change in the average speed 
of traffic, but the peak of congestion was displaced by 
about 15 minutes. 

These shifts suggest that during the months when 
traffic on the bridge was declining, some travelers were able 
to begin their morning journeys somewhat later. But the 
shifts can't be attributed definitely to BART. They might 
reflect seasonal variations in traffic behavior, or the impact 
of the 1974 gasoline shortage on travel habits. 


7 Interpretation 

Comparing BART's Performance with Expectations 

Although BART has become an important resource 
for commuters - especially in the transbay travel cor- 
ridor — it hasn't fulfilled some early predictions about its 
performance and patronage. There are two major reasons 
for this. 

First: Those predictions assumed that BART would 
provide an extensive schedule of reliable service, and 
that it therefore would compete with the private automo- 
bile for many kinds of trips. BART's operating problems, 
described in Section 2 of this report, have prevented its 
attaining those service goals. Several more years will elaspe 
before most of the problems can be solved. 

Second: Some of the original expectations about 

BART were quite unrealistic. Many people - professional 
planners, government officials and members of the general 
public — clearly anticipated more benefits than BART 
could have provided. 

BART plays a limited and specialized role in the 
transportation system of the Bay Area: It carries com- 
muters on relatively long trips between their suburban 
homes and their workplaces in central cities. This was 
BART's original purpose (as earlier sections have empha- 
sized), and it has influenced many aspects of BART's 
design and operation. 

But commuting trips, by all modes of transportation, 
represent only 30% or so of all the trips made in BART's 
greater service area. Though BART might capture a sub- 
stantial fraction of these trips, as indeed it has done, its 
impact on total travel in the region could never appear 
to be large. 

Future Patronage Growth 

Sources of New Patronage If BART's impacts on trans- 
portation and travel are to increase at all, its patronage 
must grow. Patronage now is limited by BART's unre- 
liability, by its consequently unpredictable travel times, 
and by the inadequacy of the parking lots and the local 
bus service at some of its stations. BART now is working 
actively to relieve these constraints. If it succeeds, it doubt- 
less will attract new patrons from the large group of travel- 
lers who already recognize its potential value but can't 
accept its drawbacks. 

BART also will gain new patronage through the 
growth of population and employment in its service cor- 
ridors (especially the transbay corridor) and through in- 
creases in the stresses and costs of automobile travel. As 

traffic congestion worsens and as gasoline prices, parking 
fees or bridge tolls rise, some persons who now ride in 
automobiles will turn to BART, although their BART 
trips might take more time than the corresponding auto 

Limits on Patronage Growth Even if BART improves 
its service, its reliability and access to its stations, three 
factors will act — separately and in combination — to 
restrict its patronage and its share of the travel market. 
These factors will be the relatively slow growth of the travel 
market itself, travellers' continuing dependence on the 
automobile, and constraints on BART's own capacity. 

Slow Growth of the Travel Market Projections of 
recent trends in the Bay Area indicate that population 
(illustrated in the graph below) and total employment 
will grow more slowly in the three BART counties than 
in the six other counties. 

In 1975, the BART counties had about half of the 
population and half of the jobs in the Bay Area. While 
these fractions will decline only slightly in the years be- 
tween now and 1985, the BART counties will capture no 
more than 40% of the Bay Area's total increases in pop- 
ulation and jobs during the same years. 

1950 1960 1970 1980 

Population growth in the Bay Area. During the past eight years, 
the population of the three BART counties (broken black line) 
has been almost static while that of the six non-BA RT 
counties I solid black line) has grown steadily. The colored 
line shows the population of the BART counties as a per- 
centage of the total population of all nine Bay Area counties. 


The relatively slow growth of the BART counties 
will directly affect BART's potential travel market. The 
extension of BART into other, faster-growing markets 
isn't likely to occur in the near future. 

Continuing Dependence on the Automobile Though 
automobile travel might become more difficult and more 
costly in the future, this fact remains: During the past 
30 years, most of the Bay Area's growth has occurred in 
sprawling, low-density suburbs that were designed and 
located for persons who would use automobiles for shop- 
ping, for commuting to work and to school, and for most 
other travel purposes. 

The graph in the next column reflects the importance 
of the automobile to residents of such communities. As 
population has grown in the three BART counties, the 
number of registered autos has grown even faster; and the 
number of miles of roads has increased consistently. 

No rail rapid-transit system, by itself, can greatly 
affect the travel habits of people who live in these auto- 
mobile-oriented communities. Saying that in another way: 
71 miles of BART trackways won't alter significantly — and 
mustn't be expected to alter significantly — the use of the 
6,500 miles of highways and streets in BART's greater 
service area. 

Technical Constraints on Capacity BART's present 
efforts to reduce minimum headways to 3 or 4 minutes 
might succeed during the next few years. Such success 
would approximately double BART's capacity in certain 
corridors. But as headways approached 3 minutes, the 
availability of cars would become a strong constraint 
on capacity. BART's fleet now has 450 cars; expanding 
it would be a slow process because of the expense of new 
cars and the long lead time required for purchasing them. 

If BART attempted to reduce its minimum headways 
to less than 3 minutes, it would confront further con- 
straints imposed by its automatic train-control system, by 
the capacity of its turn-around facilities at the Daly City 
terminal, and by the braking ability of its trains when they 
are travelling at high speeds. The need for adequate braking 
distances between trains would limit the minimum at- 
tainable headways. 

Together, these technical constraints suggest that 
BART might never achieve the 1 Vi-minute headways en- 
visioned in its original design objectives. 

BART's Future Role 

BART's future importance will derive from its spe- 
cialized role within the Bay Area's total transportation 
system. Its role will be influenced by events and decisions, 
both public and private, in other parts of that system. 

Peak-period travel in the Bay Area always will present 
problems because travellers must move in topographically 
restricted corridors. These problems are growing worse 
because employers are continuing to create new jobs in 
San Francisco while workers are choosing to live in the 

Regardless of BART's success in attracting patrons, 
the dependence of Bay Area residents on automobiles is 
a regional problem that will yield only to a broad, regional 
solution. This solution will have to include closer co- 
ordination between development decisions and trans- 
portation decisions, and continued expansion of all public- 
transit alternatives. BART can be only one element of such 
a solution. 

BART's major contribution will be the meeting of 
increased travel demand in the corridors that it serves, 
particularly in the transbay corridor and the corridor along 
California Highway 24. Major highway expansions in these 
corridors are unlikely, for both environmental and finan- 
cial reasons. 

BART provides essential flexibility for contending 
with future transportation needs - flexibility for trans- 
portation planners in dealing with increased travel demand, 
and flexibility for individual commuters and other trav- 
ellers in choosing the modes of transportation that serve 
them best. 

Population, automobiles and roads in the BART counties. 
The growth of population (broken black line) has been ac- 
companied by proportionally faster increases in the number 
of registered autos (solid black line) and the number of road 
miles (colored line). Data on cumulative miles of roads in the 
BART counties weren't available for the years before 1960. 


8 Technical Literature 

These MTC publications provide further information 
about the BART Impact Program and the TSTB Project. 
All of them are available from the National Technical 
Information Service (Springfield, Virginia). 

Reports by the TSTB Project 

BART's First Five Years: Transportation and Travel 

Report No. DOT-BIP-FR 11-3-78 

This is the final report of the Transportation 
and Travel Behavior Project. 

A Review of Some Anticipated and Observed Impacts 
of the Bay Area Rapid Transit System 
Report No. DOT-BIP-PD 3-1-74 

Travel in the BART Service Area 
Report No. DOT-BIP-WP 35-3-77 

Other Reports by the BART Impact Program 

Environmental Impacts of BART: Final Report 
Report No. DOT-BIP-FR 7-4-77 

Impacts of BART on Bay Area Institutions and 
Life Styles: Final Report 
Report No. DOT-BIP-FR 10-6-77 

The Economic and Financial Impacts of BART: 
Final Report 

Report No. DOT-BIP-FR 8-7-77 

BART Impacts on Highway Traffic and Transit 

Report No. DOT-BIP-TM 20-3-76 

Transportation and Travel Impacts of BART: Interim 

Service Findings 

Report No. DOT-BIP-FR 6-3-75 

Explanatory Modeling of Transbay Travel Choice 
Report No. DOT-BIP-WP 34-3-77 

Analysis of BART Capital Costs 
Report No. DOT-BIP-WP 40-3-77 

BART Impacts on Travel by Ethnic Minorities 
Report No. DOT-BIP-WP 57-3-78 

1977 Work Travel Survey Methods and Findings 
Report No. DOT-BIP-WP 58-3-78 

The Provision and Use of the BART Facilities for 
Disabled Persons 

Report No. DOT-BIP-WP 43-11-77 

The Impact of BART on Public Policy: Final Report 
Report No. DOT-BIP-FR 13-8-78 

The Impact of BART on Land Use and Urban Devel- 
opment: Final Report 
Report No. DOT-BIP-FR 14-5-78 

BART in the Bay Area 

Report No. DOT-BIP-FR 9-201-78 

This is the final report of the BART Impact 
Program. It summarizes and intergrates the 
findings of the program's several projects.